Abstract
Traditional agency theory views the proper role of the board chair exclusively as providing independent oversight to monitor and control the CEO. Recently, firms have introduced innovations in board leadership that have confounded these theoretical expectations. One notable innovation is the executive board chair, a corporate governance hybrid responsible for both oversight and strategic decision-making, challenging agency theory's prescription that the two activities remain separate. In this study, we argue that an executive board chair position can resolve the trade-off between independent oversight and involvement in strategy and therefore generate a performance advantage. We also predict that, owing to the blurring of lines between the CEO and board chair roles that the executive board chair position creates, the relationship will be stronger the greater the need to monitor and control the CEO but weaker when organizational complexity and board leadership demands are greater. Analysis of S&P 1500 firms from 2003 to 2017 provides general support for our arguments.
The role of the board of directors in the modern corporation is complex and ambiguous. The board must simultaneously monitor and oversee the firm's executives, while also collaborating with those same executives on firm strategy (Sundaramurthy & Lewis, 2003). Perhaps the most visible and controversial manifestation of this ambiguity is the question of board leadership: Who shall chair the board? For decades, research on board leadership was limited to investigating the relative merits of the CEO also serving as board chair, a practice known as CEO duality (Krause, Semadeni, & Cannella, 2014). However, as firms have increasingly separated their CEO and board chair positions (Spencer Stuart, 2019), researchers have begun to develop newer, more nuanced theory around the unique role the board chair performs when separate from the CEO (e.g., Hoppmann, Naegele, & Girod, 2019; Krause, 2017; Withers & Fitza, 2017). Scholars have demonstrated that board chairs significantly impact their firms, acting as a resource (Krause, Semadeni, & Withers, 2016b), driving strategic change (Hoppmann et al., 2019), determining director engagement (Bezemer, Nicholson, & Pugliese, 2018), and ultimately explaining a significant amount of variance in firm performance across institutional contexts (Krause, Li, Ma, & Bruton, 2019; Withers & Fitza, 2017).
Recently, boards have introduced corporate governance innovations that have confounded extant theoretical conceptualizations of board leadership (Krause, Withers, & Semadeni, 2017; Semadeni & Krause, 2020). One such innovation is the executive board chair position. The executive board chair leads the board in its oversight responsibilities and is separate from the CEO, consistent with the prescriptions of agency theory. However, unlike a non-executive board chair, the executive chair also leads the firm's strategic decision-making and may even be involved in its implementation. While these additional responsibilities raise questions about the executive board chair position's efficacy in monitoring, many have argued that effective oversight requires in-depth knowledge of firm activities (e.g., Baysinger & Hoskisson, 1990; Zorn, Shropshire, Martin, Combs, & Ketchen, 2017) and that firms can benefit from the strategic resources a chair offers (Krause et al., 2016b; Withers & Fitza, 2017). Thus, it may be that mixing oversight and strategic decision-making, as represented in the executive board chair position, benefits the firm's governance. However, the consequences of the executive board chair structure for firm performance remain wholly unknown, and with nearly a quarter of all separate board chairs in the S&P 500 now designated as “executive” (Spencer Stuart, 2019), knowledge of these consequences has never been more important.
In this study, we address the theoretical complexities of this emerging governance practice by conceptualizing the executive board chair position as a structural choice and examining its impact on within-firm variance in firm performance. Drawing on prior literature, we argue that the executive board chair's closer involvement in strategic decision-making and implementation will enable more effective monitoring and control of the firm's management and a greater use of the chair's strategic knowledge and resources. Based on this logic, we predict that a given firm will have higher average performance with an executive board chair than with another type of board chair. As the primary mechanism of this relationship is more informed monitoring from board chair involvement in strategic decision-making, we argue that the relationship will be stronger the greater the need to monitor and control the CEO. Further, as the executive chair is a hybrid governance position, responsible for both strategy and board leadership, we also argue that the position may blur the traditional boundaries of the CEO and board chair positions. We predict that when a firm's organizational complexity or board leadership demands are greater, the lack of role clarity of the CEO and chair positions created by the blurring of these boundaries may cause the benefits of the position to diminish. We test our hypotheses using within-firm analysis of 2,162 firm-year observations from 289 firms listed in the S&P 1500 index that were led by an executive board chair at some point from 2003 to 2017. Results reveal general support for our hypotheses. Specifically, we find that a given firm will have higher performance with an executive board chair than with another type of board chair. This effect is stronger when a firm's CEO is more powerful but weaker when a firm's organizational complexity or board leadership demands are greater.
We seek to contribute to theory and practice with this research. We contribute to the ongoing debate within agency theory about the role of insider knowledge and experience in the board's control and monitoring functions (Baysinger & Hoskisson, 1990; Ocasio, 1994; Shen & Cannella, 2002; Zorn et al., 2017) by introducing a hybrid board leadership role rarely discussed in the literature but frequently used in practice. The executive board chair mixes strategy and oversight responsibilities in a way that a non-executive board chair or a combined CEO/chair cannot. By revealing that a firm is likely to exhibit higher performance with an executive board chair than with another type of board chair, we provide evidence that this position may help to resolve the trade-off between involvement and independence in governance. However, we also delineate some boundary conditions to this relationship, noting that the nature of the executive board chair position has implications for duties normally entrusted to the CEO and board chair. In doing so, our research also contributes to practice by further aligning theory with the changing landscape of corporate governance and offering evidence of a performance advantage for a firm that decides to innovate its governance beyond the dichotomous constraints of CEO duality (Semadeni & Krause, 2020).
Theory and Hypotheses
The literature on board leadership presents the board chair role as being split into three main responsibilities: monitoring, strategic advice, and board leadership. First, according to formal agency theory, the board of directors plays an important role in addressing and mitigating the potential for principal-agent conflict between a firm's owners and its managers, respectively (e.g., Eisenhardt, 1989; Fama & Jensen, 1983; Jensen & Meckling, 1976). As the head of the board, the chair has been considered a key player in leading the board's monitoring activities, ensuring that management acts in the best interests of the firm's shareholders (e.g., Berg & Smith, 1978; Rechner & Dalton, 1991). Second, the board of directors also plays an important role in firm strategy (Pugliese et al., 2009) and the board chair's role includes strategic advice and guidance to the CEO. This view posits that the chair does not simply monitor management but also aids the CEO with strategic decision-making (Krause, 2017; Krause et al., 2016b; Withers & Fitza, 2017). Third, the board of directors is a team composed of directors with different interests and loyalties (Johnson, Schnatterly, & Hill, 2013; Ocasio, 1994; Tuggle, Schnatterly, & Johnson, 2010). Accordingly, the literature views the chair's board leadership role as requiring considerable efforts in managing directors’ engagement and interactions in order to ensure healthy board functioning essential for effective oversight of, and advice to, management (Bezemer et al., 2018; Hoppmann et al., 2019; Veltrop, Bezemer, Nicholson, & Pugliese, 2021).
Considering these responsibilities, scholars have for decades debated the pros and cons of separating the board chair position from the CEO position. Some have noted that having the CEO also lead the board in performing its duties challenges the assumption that the board is willing or even able to exercise independent oversight of management, particularly with regard to the CEO (Rechner & Dalton, 1991). Some also suggest that a non-CEO chair can bring additional resources to aid a firm and its CEO in strategy (Krause et al., 2016b; Withers & Fitza, 2017). Conversely, others suggest that if the board remains diligent in its monitoring efforts, it may not necessarily need the two roles to be held separately (Finkelstein & D’aveni, 1994) and that combining the chair and CEO positions aids in unity of direction (Fayol, 1949), which allows the CEO to implement strategy more efficiently and effectively (Boyd, 1995; Donaldson & Davis, 1991; He & Wang, 2009). The diligence of the board in both monitoring and strategic guidance, however, is something that must be continually cultivated, requiring a board chair to dedicate time and energy to developing directors and managing their interactions (Bezemer et al., 2018; Hoppmann et al., 2019; Veltrop et al., 2021), something that a non-CEO chair—who is typically not expected to lead a firm in its strategy development and implementation—should be in a better position to do. Accordingly, the choice between combining and separating the chair and CEO positions is often considered to be a double-edged sword, each side with its own advantages and disadvantages (Finkelstein & D’aveni, 1994).
The executive board chair represents a peculiar board leadership approach because the executive chair fulfills the responsibilities of the non-CEO board chair in monitoring and board leadership while still remaining actively involved in strategy development and implementation. Executive board chairs “are employees of the companies by definition and take more active roles in supporting the CEO's leadership of the company” (Bradt, 2013). Our conceptualization of the executive board chair position builds on an inspection of the press releases from the S&P 1500 firms in our sample announcing the appointment of an executive board chair. We carried out an exploratory qualitative analysis of those firms that described the roles of their executive board chairs, from which three main roles emerged: 70 percent described that, beyond performing the typical oversight duties of a non-CEO board chair, the executive board chair would be directly involved in strategic decision making and planning; 19 percent mentioned that the executive board chair would focus on investor and other external stakeholder relationships; and 17 percent mentioned a focus on business development efforts. For example, Haliburton (2017) stated that its executive chair would “play an important leadership role focussing on the strategic direction of the company … be actively engaged with shareholders, and continue working with customers to ensure the Company is best addressing their needs.” Spartan Nash (2008) adopted a more internal focus, stating that its executive board chair would “manage board functions and facilitate interaction between the board and executive management” as well as “coordinate with the CEO to develop and execute the Company's business strategy, cultivate a performance-driven corporate culture, mentor executive leadership, and assist in maintaining critical business relationships.”
Such statements were common among the press releases we examined, suggesting that the executive board chair acts as the strategic leader of the organization as well as the leader of the oversight body, blurring the line of demarcation between the board and management. This is in line with research that has expanded the board's theoretical role past that of solely monitoring to include advice and guidance, positing that boards often are more involved in strategic decision-making than traditional agency theory would suggest (Golden & Zajac, 2001; Hillman & Dalziel, 2003). From this perspective, the executive board chair enables a firm to better exploit its chair's resources and leverage their human and social capital to improve strategic decision-making (Krause et al., 2016b), while still maintaining effective oversight.
Table 1 provides an overview of the three types of board chair positions and their respective responsibilities. As evident, a non-executive board chair performs a limited strategic role and is instead focused more on monitoring and board leadership activities. Conversely, a CEO/chair is indeed the CEO, focused on strategy development and implementation, thus playing little to no role in oversight, and a constrained role in board leadership; these responsibilities fall to the independent directors on the board, now typically led by a lead independent director (Dalton, Hitt, Certo, & Dalton, 2007; Krause et al., 2017). However, an executive board chair has high responsibilities over all three roles. As a non-CEO board chair, the executive chair holds responsibilities in monitoring and board leadership activities; as an executive officer of the firm, the executive chair holds responsibilities in strategy development and implementation. The executive board chair, then, is a hybrid governance phenomenon. This raises an important question: What are the performance implications for a firm opting for an executive board chair?
Board Chair Structures and Responsibilities
The Executive Board Chair and Firm Performance
We argue that, ceteris paribus, the greater involvement of the board chair in firm management should enhance the chair's ability to lead the board in performing its corporate governance responsibilities and at the same time aid the CEO in strategic decision-making and implementation. This, we argue, should improve a firm's performance. Our argument builds on the long-running debate among governance scholars regarding the role of firm-specific knowledge and strategic involvement in board governance activities (e.g., Baysinger & Hoskisson, 1990; Rindova, 1999; Zorn et al., 2017). For decades, agency theorists maintained that boards should be composed predominantly of independent outsiders, based on the assumption that executives of the firm cannot objectively evaluate their own decisions and performance (e.g., Fama & Jensen, 1983; Mallette & Fowler, 1992). Though CEOs generally serve on the board in most institutional contexts, governance experts have long advocated for the exclusion of other executives, as these executives are assumed to be beholden to the CEO, and thus ineffective as a source of vigilant governance (e.g., Monks & Minow, 2008; Weisbach, 1988). Only under specific circumstances, usually when the CEO is already weakened, can other inside directors reasonably be expected to contribute to the board's oversight of their boss (Ocasio, 1994; Shen & Cannella, 2002).
Nevertheless, many scholars have noted a problematic trade-off associated with the predominance of outside directors: Insiders possess far superior knowledge about the firm and its performance, potentially making them useful governance assets (e.g., Baysinger & Hoskisson, 1990; Ocasio, 1994; Rosenstein & Wyatt, 1997; Zorn et al., 2017). As firms have steadily removed non-CEO executives from their boards following passage of the 2002 Sarbanes-Oxley Act (Joseph, Ocasio, & McDonnell, 2014), researchers have shown that a board's ability to govern the CEO declines when the CEO is the only insider (Zorn et al., 2017). As an inside director, the executive board chair reduces information asymmetry between the CEO and the board, and this position's mere presence is likely to put more performance pressure on the CEO (Baysinger & Hoskisson, 1990; Zorn et al., 2017).
Of course, agency theorists’ criticism of inside directors is not without merit; it is unlikely that the CEO's subordinates will be able to contribute much to board governance if it means challenging the CEO. An executive board chair, however, faces no such hierarchical impediments. An executive board chair possesses structural power over the CEO by virtue of their title (Finkelstein, 1992) and thus faces little to no threat of reprisal should they exercise control or monitoring over the CEO in their capacity as chair. Thus, when a board has an executive board chair, it can experience the benefits of both strong oversight and strategic knowledge; an executive chair has better knowledge with which to govern than a non-executive chair and greater propensity to lead the board in exercising governance over the firm's management than a combined CEO/chair.
At the same time, the executive board chair's greater involvement in strategic decision-making and implementation also enables the firm to benefit from their knowledge and resources. Board chairs are normally highly experienced executives who bring with them useful human capital that can help in the running of the firm (Krause et al., 2016b; Shekshnia, 2018). A close collaboration between the executive board chair and CEO may offer a benefit to firm performance by exploiting the chair's knowledge and resources (Krause, 2017). When boards have the ability to influence strategic decisions, they tend to do so (Boivie, Withers, Graffin, & Corley, 2021; Garg & Eisenhardt, 2017; Golden & Zajac, 2001). An executive board chair may also serve as a more effective liaison between the board and CEO (Bezemer et al., 2018), enabling greater access to the rest of the board's resources and knowledge as well (Hillman & Dalziel, 2003), thus offering further advantages.
In short, through greater involvement in strategic decision-making and implementation, the executive board chair position can potentially provide the firm with effective oversight of management as well as the ability to leverage the board chair's knowledge and resources. We posit that, as a result, when a firm has an executive board chair, it will perform better than when it has a different type of board chair.
Hypothesis 1: A given firm will have higher performance with an executive board chair than with another type of board chair.
The hybrid nature of the executive board chair position means that, to some extent, the boundaries of the CEO and board chair roles are blurred. Unlike a non-executive board chair, the executive board chair actively engages in strategic decision-making and implementation, therefore assuming some of the roles traditionally entrusted to the CEO. And yet as a separate chair, the executive board chair is still the leader of the board the way that a CEO/chair (whose governance duties are assumed by a lead independent director) often is not. In the following sections, we argue that the blurring of these boundaries has implications for how effective the executive chair position may be in each of the three duties expected of a separate board chair.
The Executive Board Chair and Oversight of Management
As we have suggested, the executive board chair assumes some authority over strategic decisions, a responsibility normally entrusted to the CEO. Owing to this blurring of responsibilities of the two positions, one of the benefits of the executive board chair position is closer monitoring of the CEO. Hence, the greater the need to monitor the CEO, the greater the benefit the executive board chair position can offer a firm. Extensive research in the corporate governance literature indicates that CEO power is a crucial contingency factor in the board's ability to monitor and control the CEO (e.g., Daily & Johnson, 1997; Krause et al., 2017; Ocasio, 1994; Shen & Cannella, 2002; Zajac & Westphal, 1996). One of the major sources of CEO power is, in fact, the board chair title. However, CEOs derive power from several other sources, including stock ownership, tenure in the CEO position, and compensation, among others (Daily & Johnson, 1997; Krause, Filatotchev, & Bruton, 2016a; Krause, Priem, & Love, 2015).
Initially, agency theorists prescribed filling the board with outside directors as the way to ensure independent and objective oversight of the CEO (e.g., Fama & Jensen, 1983; Weisbach, 1988). However, many scholars have argued that even independent boards vary in the power they possess relative to the CEO, with some CEOs commanding enough power to sway boards to their point of view even without any formal hierarchical authority (e.g., Daily & Dalton, 1994; Westphal & Zajac, 1995; Zajac & Westphal, 1996). In addition, an independent board has less firm-specific information, which can lead to greater difficulty monitoring a powerful CEO's behavior (Baysinger & Hoskisson, 1990; Zorn et al., 2017). As leader of the board and primary liaison between the board and CEO, the chair serves an important role in a firm's oversight mechanism. Accordingly, the more effective the chair is as a monitor, the greater the benefit to the firm.
An executive board chair can prove a significant asset if the board looks to its chair to act as a counterweight to a powerful CEO. Because an executive board chair moves past simply monitoring and advising the CEO and takes on an active role in the management of the firm, an executive chair usurps some of the CEO's unilateral authority over strategic decisions, lowering the dangers of entrenchment activities. When a CEO is less powerful, mitigating the information asymmetries that can lead to entrenchment activities is easier because the CEO's influence over management and the board is lower (Shen, 2003). Indeed, research shows that a chair is easily able to impose his or her own will over a less powerful CEO (Quigley & Hambrick, 2012). In such a case, the benefit that an executive chair offers the firm may be less pronounced. As a CEO gains power, however, monitoring becomes more challenging for an outsider-dominated board because the CEO controls the flow of information to the board. Because the executive board chair is actively involved in strategic decision-making, he or she can provide the firm-specific information the board needs to counter a powerful CEO's influence. Hence, when a firm has an executive board chair, it is likely to exhibit a greater performance advantage when its CEO is more powerful, and the risk of information asymmetry is greater, than when its CEO is less powerful.
Hypothesis 2: A given firm's performance benefit from having an executive board chair (relative to another type of board chair) is higher when CEO power is higher.
The Executive Board Chair and Involvement in Strategy
Though the blurred structural lines of an executive board chair's involvement in both monitoring and strategy may offer benefits, this ambiguity may, at times, also risk impeding a CEO's ability to carry out firm strategy effectively. We argue that this may be the case when the firm is more complex. The complexity of a firm can increase as a firm increases in size, (related) diversification, or acquisition activity (Donaldson, 2001; Hambrick & Cannella, 2004; Haspeslagh & Jemison, 1991; Josefy, Kuban, Ireland, & Hitt, 2015). The greater difficulty in coordinating between business units, customers, suppliers, and other stakeholders that organizational complexity entails requires that knowledge processing and decision-making is streamlined and consistent (Boyd, 1995; He & Wang, 2009; Hill, Hitt, & Hoskisson, 1988; Mintzberg, 1973). A failure to manage the increased knowledge and decision-making requirements associated with greater organizational complexity may result in a breakdown in strategy execution, ultimately harming the firm's performance.
The challenge that an executive board chair poses to managing organizational complexity is that there are essentially two strategic leaders of the firm, weakening what Fayol (1949) called “unity of direction.” When a CEO's unity of direction is weakened by another strategic leader, it hinders the CEO's ability to unilaterally carry out firm strategy; the result is a lack of consistency in strategic decision-making that can hinder strategic change and harm firm performance (Krause et al., 2015; Quigley & Hambrick, 2012; Zhang, 2006). The more complex the firm becomes, the more important it is that strategy be developed and executed consistently (Connelly, Tihanyi, Ketchen, Carnes, & Ferrier, 2017). When an organization is less complex, strategy is more straightforward and conflicting strategic directives should be less common and less likely to adversely impact the organization, as executives face lower discretion overall and there are only so many directions along which strategic leaders could diverge (Hambrick & Abrahamson, 1995; Hambrick & Finkelstein, 1987). As the organization becomes more complex, however, strategic decisions become more complex, and disagreements between strategic leaders at the top of the firm are more likely (Carpenter, 2002). Such disagreements will have the potential to sow discord and confusion among a firm's managers and their subordinates, harming firm performance (Krause et al., 2015).
Typically, the CEO position represents the highest ranking executive officer of the firm, with all other executive positions being subordinate to it (Finkelstein, 1992). The executive board chair position disrupts this pattern by blurring the lines of strategic hierarchy. The executive chair's close involvement in strategic decision-making and implementation means that another strategic leader must also receive and process knowledge, and come to decisions on strategic matters. The executive chair thus, to some extent, complicates the traditional decision-making hierarchy and restricts the CEO's authority to manage and delegate strategy execution unilaterally and consistently. This differs from either of the two alternative board leadership structures: A CEO/chair is entrusted with considerable unity of command and left to develop and implement a firm's strategy (Donaldson & Davis, 1991; Finkelstein & D’aveni, 1994); similarly, a non-executive board chair leaves the management of the firm's strategic initiatives to the CEO, instead focusing more on oversight and board leadership duties (Banerjee, Nordqvist, & Hellerstedt, 2020).
The benefits of an executive board chair's involvement in strategy should therefore wane as a given firm becomes increasingly complex. When a firm's organizational complexity is low, the blurred lines between strategy and oversight created by an executive chair are less likely to challenge unity of direction, as strategic direction is a more straightforward endeavor (Carpenter, 2002). The firm should be able to function well with the added involvement of an executive chair. As the firm becomes more complex, unity of direction becomes harder to achieve with two strategic leaders guiding firm decisions (Krause et al., 2015). This should reduce the overall benefit that the executive board chair position offers the firm.
An alternative perspective could be applied to the question of organizational complexity that would yield the opposite prediction. Specifically, if we assume that the executive board chair predominantly acts as a strategic adviser and sounding board for the CEO, then it should follow that the benefits of the executive chair would increase when a firm is more organizational complex. However, in contrast to a non-executive chair or another outside director or even a former CEO acting as adviser, the executive chair has not only hierarchical authority over the CEO but is generally viewed as a strategic leader of the firm as well. An adviser or other board member could provide input on how to manage complexity without that input adversely impacting the consistency of strategic direction communicated throughout the firm. An executive chair, however, cannot provide advice without the added complication of hierarchical authority. As such, we expect that the potential for disagreement and loss of unified direction among the two strategic leaders when the firm is more complex will outweigh potential benefits from combining multiple perspectives.
Hypothesis 3: A given firm’s performance benefit from having an executive board chair (relative to another type of board chair) is lower when organizational complexity is higher.
The Executive Board Chair and Board Leadership
The hybrid nature of the executive board chair position may also blur the lines between the executive chair's strategic responsibilities and their responsibilities as group leader of the board. As the board chair of SI International Inc. stated, “The role of chairman has become a full-time job, given the increased duties and responsibilities associated with being a public company” (Gerin, 2005). Indeed, a chair must manage directors’ engagement, interests, and interactions to ensure proper functioning of the board (Bezemer et al., 2018; Hoppmann et al., 2019; Veltrop et al., 2021). This can include keeping directors’ personal interests at bay, promoting self-evaluations and professional development of directors, and improving board functioning practices and routines (Hoppmann et al., 2019). A chair's duties also include preparing for meetings and understanding directors’ backgrounds and priorities in order to ensure that all contribute to the board in a healthy and effective way (Bezemer et al., 2018). A successful board chair is able to develop a more attentive board (Tuggle, Sirmon, Reutzel, & Bierman, 2010) and manage disagreements between directors and the CEO (Veltrop et al., 2021). Board leadership duties may vary depending on the number of directors, how busy they are, and how many of them are new. These board leadership demands add to the chair's workload, requiring a greater amount of the chair's time and effort (Fich & Shivdasani, 2006; Hambrick, Finkelstein, & Mooney, 2005; Hoppmann et al., 2019; Lipton & Lorsch, 1992; Veltrop, Molleman, Hooghiemstra, & van Ees, 2018).
The emphasis on strategy inherent in the executive board chair position may result in a lack of clarity around expectations regarding board leadership duties. This differs from either of the alternative board leadership structures. A CEO/chair is not expected to lead the board in its oversight and strategic advisory duties as the CEO would essentially be leading their own advising and monitoring. Nor is a CEO/chair responsible for managing and developing directors and their engagement in these responsibilities. Rather, when the firm has a CEO/chair, the board clearly undertakes these responsibilities itself (Finkelstein & D’aveni, 1994), often appointing a lead independent director to lead the board in these tasks (Krause et al., 2017). Conversely, the non-executive board chair separates governance from strategy to ensure independent oversight of management's strategic decision-making and execution (Rechner & Dalton, 1991). As such, the non-executive chair's role is to focus largely on duties related to healthy board functioning.
But as we have mentioned, the executive board chair position holds a special responsibility for strategy. In certain circumstances, this emphasis on strategy may result in an executive chair's failure to properly address board leadership duties. When the demands of such duties are low, addressing them should be easier. Moreover, any detriment to the firm of not fully doing so should be low. However, when a given firm's board leadership demands are greater, an executive board chair may be even less likely to address them adequately (Hambrick et al., 2005), as their focus will be split with strategic matters. As the needs of board leadership increase, the failure to fully address them will do increasingly more harm and cause a deterioration of board functioning. This can lead to increased agency costs (Jensen & Meckling, 1976), strategic inertia (Hoppmann et al., 2019), conflict among board members (Bezemer et al., 2018; Veltrop et al., 2021), and the departure of quality directors (Garg, Li, & Shaw, 2018), all of which harm firm performance.
Hypothesis 4: A given firm’s performance benefit from having an executive board chair (relative to another type of board chair) is lower when board leadership demands are higher.
Methods
We test our hypotheses on a sample of S&P 1500 firms between the years 2003 and 2017, with firm performance data collected through 2020. We followed recent work in corporate governance and chose 2003 as the first year of our sample, since the passing of the 2002 Sarbanes-Oxley Act fundamentally changed many aspects of corporate governance (Krause & Semadeni, 2013; Krause & Semadeni, 2014). Board and executive data were collected from the BoardEx and Execucomp databases. Firm-level data were obtained from Compustat with the exception of acquisition data, which were collected from SDC's Mergers and Acquisitions database. After accounting for missing data, our initial sample consisted of 10,420 observations from 1,545 firms. This sample was used in our first-stage sample selection analysis described below. Since the tests of all hypotheses were based on within-firm analysis, and because of the necessity for variance in executive board chair presence, we limited our final sample to firms that had an executive board chair at some point between the years 2003 and 2017. This final sample consisted of 2,162 observations from 289 firms.
Measures
Dependent variable
Our theory suggests that the executive board chair's influence on firm performance comes from being more closely involved in strategic decision-making and implementation. This should be most visible in an accounting-based measurement of firm performance. As such, our dependent variable of firm performance is operationalized as a firm's return on assets (ROA). This variable is calculated as a firm's annual net income divided by its total book value of assets. We test performance over several time periods, examining one-year, two-year, and three-year ROA. One-year ROA is calculated as a firm's ROA at year t + 1; two-year ROA is calculated as the yearly average of a firm's ROA between t + 1 and t + 2; three-year ROA is calculated as the yearly average of a firm's ROA between t + 1 and t + 3.
Independent variables
We measured executive board chair as a binary variable taking the value of “1” if a firm had an executive board chair in year t and “0” otherwise. This was determined by manually examining whether a firm's board chair held the title of “executive (board) chair” as reported by the BoardEx database. For the manual inspection, we considered the few variations of the executive board chair title, including “executive chair (wo)man” and “executive chairperson.” Given that some firms have an executive board chair temporarily following a CEO succession, we required that an executive board chair have a tenure of at least 1 year in order to be designated as “1.” All cases in which an executive board chair held the position for less than 1 year were given a “0”.
Following recent studies, we measured CEO power as an index of multiple measures in order to capture the multidimensional nature of the power construct (Cannella & Shen, 2001; Krause et al., 2016a; Krause et al., 2015). Our index of CEO power is the sum of the standardized values of CEO tenure, CEO salary, and CEO stock ownership, all collected from the Execucomp database. CEO tenure, measured as the number of years the CEO has held the position, can be an important source of power and authority as a CEO's influence over the firm and board may increase with tenure (e.g., Cannella & Shen, 2001; Hill & Phan, 1991; Van Essen, Otten, & Carberry, 2015). CEO salary, which excludes bonuses or other incentive pay, has regularly been considered indicative of CEO power (Daily & Johnson, 1997; Finkelstein, 1992; Krause et al., 2015). CEO stock ownership, the proportion of the firm's outstanding shares owned by the CEO, is also considered to be related to the CEO's ability to influence firm decisions (Cannella & Shen, 2001; Finkelstein, 1992; Krause et al., 2015).
Prior research has suggested that firms that are larger, whose related diversification profile is greater, and who engage in more acquisitions have greater organizational complexity (Hambrick & Cannella, 2004; Menz & Scheef, 2014). Therefore, we measured organizational complexity as an index of the sum of the standardized values of firm size, related diversification, and acquisition activity. Firm size was collected from the Compustat database and calculated as a firm's number of employees in the focal year. Related diversification was collected from the Compustat database and calculated with Palepu’s (1985) entropy measure based on the firm's dispersion of sales across 4-digit standard industrial classifications (SICs) within the firm's 2-digit SIC in the focal year (Hambrick & Cannella, 2004). Acquisition activity was collected from the SDC's Mergers & Acquisitions database and calculated as the number of acquisitions in the focal year that were at least 2% the size of the focal firm (Hambrick & Cannella, 2004).
Extant research has also noted that boards with more members (Jensen, 1993; Lipton & Lorsch, 1992), more members who sit on additional boards (i.e., busy directors; Fich & Shivdasani, 2006), and more new members (e.g., Veltrop et al., 2018) have greater board leadership demands on the board chair. We therefore measured board leadership demands as the sum of the standardized values of board size, the proportion of busy directors, and the proportion of new directors. These variables were collected from the BoardEx database. Board size was calculated as the total number of directors who sat on the board. A busy director was considered as such if the director sat on three or more external boards (Fich & Shivdasani, 2006). A new director was considered as such if the director was in his/her first year as a director of the firm.
Control variables
We controlled for a number of firm-, board-, and executive-level factors that may affect firm performance. We considered whether the CEO was an insider CEO or not, as this can also affect firm performance (Karaevli, 2007; Quigley, Hambrick, Misangyi, & Rizzi, 2019). A CEO was considered an insider CEO if he or she was an incumbent executive of the firm prior to being appointed CEO (Boeker & Goodstein, 1993; Shen & Cannella, 2002). Extending the threshold for insider CEO status to 1 or 2 years of firm tenure prior to promotion did not change our results. We controlled for the chair's role tenure and board tenure as firms may benefit from board chairs with greater human and social capital within the firm (Krause et al., 2016b). To account for the influence of the board chair, we measured chair ownership as the proportion of the firm's outstanding shares owned by the firm's board chair. To account for a board's efforts toward independent oversight of management, we controlled for board independence, calculated as the proportion of directors who were independent directors, and whether there was a lead independent director present (Krause et al., 2017), measured with a dummy variable. Since a recent change in board chair structure can also affect firm performance (Krause & Semadeni, 2013), we controlled for the firm's board chair structure in the previous year by including one binary variable taking the value “1” if the firm had a Non-Executive Chair at t-1 and “0” otherwise, and another binary variable taking the value of “1” if the firm had CEO Duality at t-1 and “0” otherwise. Given that a CEO's power can be offset by the power of TMT members (e.g., Finkelstein, 1992; Ocasio, 1994; Zhang, 2006), we also controlled for TMT power, measured as the sum of the standardized values of the average salary, tenure, and ownership of the non-CEO executives reported by the Execucomp database. We controlled for firm research and development (R&D) intensity, measured as a firm's R&D expenditures divided by its sales (Quigley & Hambrick, 2012). We also controlled for industry instability, measured as the absolute difference in industry growth rate from t-2 to t-1 versus t-1 to t (Hambrick & Cannella, 2004). Finally, we included year dummies to capture any year effects.
Analysis
Our research question focuses on whether a given firm enjoys a performance benefit when it has an executive board chair relative to another type of board chair. Owing to the nature of our research question, we followed prior work (e.g., Fich & Shivdasani, 2006; Krause et al., 2017) and tested our hypotheses using within-firm analysis. As such, our sample consists only of firms in which an executive board chair was present at some point from 2003 to 2017, which allows us to examine firm performance up to and including the year 2020. This presents the possibility of sample selection bias because the independent variable, executive board chair, is related to sample selection. To correct for this possible source of endogeneity, we followed recommendations from Certo, Busenbark, Woo, and Semadeni (2016) and employed a two-stage Heckman selection model, predicting an observation's inclusion in the final sample in the first stage and then testing our hypotheses in the second stage (Heckman, 1979). The outcome of the first-stage model took a value of “1” if the observation was from a firm that had employed an executive board chair within the sample window and “0” otherwise.
All first-stage predictor variables are measured in year t. In order to meet the exclusion restriction, a variable must be included that predicts inclusion in the sample but does not directly predict the theoretical outcome. We used the executive board chair industry frequency at the three-digit SIC level as a variable that meets the exclusion restriction. This variable was selected following the rationale that firms whose peers have opted for an executive board chair should be more likely to do the same. However, the frequency of executive board chairs among a focal firm's peers should not be a predictor of the focal firm's performance. In the first-stage analysis, we ran a probit model to predict an observation's inclusion in the final sample. We then obtained the predicted probabilities from the model to calculate the inverse Mills ratio, which is included as a control in second-stage models when suitable. All second-stage models are linear ordinary least squares (OLS) regression models with firm fixed effects and robust standard errors.
To evaluate whether our exclusion restriction was sufficiently strong, we followed recommendations by Certo et al. (2016) and checked the correlation between our independent variable—executive board chair—and the inverse Mills ratio, as well as the pseudo R-squared from the first-stage model. The correlation between our independent variable and inverse Mills ratio was −0.65 and the pseudo R-squared from our first-stage model was 0.21. Both of these suggest that the strength of our exclusion restriction is sufficient to control for any selection bias (Certo et al., 2016). In addition, we also followed Certo et al.’s (2016) recommendations and only include the inverse Mills ratio in second-stage models when its p value is below 0.1, as not doing so might bias coefficients. However, removing it from these models did not materially affect the results.
Results
Our data reveal that the executive board chair position is becoming increasingly prevalent. Figure 1 demonstrates that the percentage of non-CEO board chairs in S&P 1500 firms that were executive board chairs increased by approximately two thirds between 2003 and 2017, now nearing 20 percent. Perhaps unsurprisingly, the executive board chair position is often held by someone with a deep understanding of the firm. An exploratory analysis revealed that in 35 percent of observations in which an executive board chair was present the executive chair was the firm's founder, in 11 percent the executive board chair was a member of the founding family, and in 40 percent the executive board chair was the firm's retired CEO who had transitioned to the role (but not a founder or member of the owning family). The average tenure of the executive board chairs in our sample was 4.4 years.
Proportion of separate Board Chairs in the S&P 1500 that were Executive Board Chairs
Table 2 provides means, standard deviations, and correlations from our final sample. The descriptive statistics reveal that 24 percent of the observations used for the within-firm analysis included an executive board chair. Of the other 76 percent, 15 percent included a non-executive board chair, and 61 percent included CEO duality. Table 3 reports the results from the first-stage model. As expected, the industry frequency of executive board chair presence is positively related to an observation's inclusion in the final sample (p = 0.000).
Descriptive Statistics and Correlations
Note: Correlations > |0.05| are statistically significant at p < 0.05. N = 2,082. This number represents the sample for which there are values for the three dependent variables, one-year, two-year, and three-year ROA.
First-Stage Probit Estimates of Sample Selection
Note: Robust p values are in parentheses.
Hypothesis Tests
To understand the strength and longevity of the within-firm relationship between the executive board chair position and firm performance, we ran tests on 1-year, 2-year, and 3-year average ROA, which are reported in Tables 4 to 6, respectively. The moderator variables—CEO power, organizational complexity, and board leadership demands—were standardized prior to the second-stage analyses in order to avoid multicollinearity. In each table, Model 1 includes the control variables; Model 2 also the main variable of interest, executive board chair; Models 3 to 5 each interaction term individually; and Model 6 all variables and interaction terms. In reporting our findings, we cite the full models including for the main effect, as the full models are the most appropriately specified since the interactions are significant and the moderator variables are mean-centered (Aguinis, Edwards, & Bradley, 2017; Busenbark, Graffin, Campbell, & Lee, 2021).
Second-Stage Fixed Effects OLS Estimates of Firm Performance: One-Year ROA
Note: Robust p values are in parentheses.
Second-Stage Fixed Effects OLS Estimates of Firm Performance: Two-Year ROA
Note: Robust p values are in parentheses.
Second-Stage Fixed Effects OLS Estimates of Firm Performance: Three-Year ROA
Note: Robust p values are in parentheses.
Model 6 in Tables 4 to 6 reveal a positive within-firm relationship between the use of the executive board chair position and one-year (p = 0.043), two-year (p = 0.001), and three-year (p = 0.002) measures of performance, respectively, thus providing support for Hypothesis 1. For 2-year and 3-year ROA, the main effect of executive board chair is significant when interactions are excluded from the model as well as included; for one-year ROA, the main effect is only significant in the fully specified model. The coefficients suggest a considerable performance benefit for a given firm when an executive board chair is present. For example, the average two-year ROA associated with an executive board chair is about 7.2 percent, compared to the two-year ROA of 5.4 percent associated with the alternative chair types.
Hypothesis 2 states that a given firm's performance benefit of having an executive board chair (relative to another chair type) is higher when a CEO is more powerful. Model 6 in Tables 4 to 6 offer general support for Hypothesis 2 for one-year (p = 0.012), two-year (p = 0.010), and three-year (p = 0.046) performance, respectively. To examine the moderating role of CEO power further, we plotted the predicted values of two-year ROA. As Figure 2 illustrates, when a given firm's CEO is more powerful, the advantage from an executive board chair position is considerably higher.
Interaction of Executive Board Chair and CEO Power
Hypothesis 3 states that a given firm's performance benefit of having an executive board chair (relative to another chair type) is lower when organizational complexity is higher. Model 6 in Tables 4 to 6 provide consistent support for Hypothesis 3 for one-year (p = 0.013), two-year (p = 0.026), and three-year (p = 0.009) performance, respectively. We also plotted the predicted values of two-year ROA for this interaction. As evident in Figure 3, when a given firm's organizational complexity is very great, the performance benefit of an executive board chair position disappears.
Interaction of Executive Board Chair and Organizational Complexity
Hypothesis 4 states that a given firm's performance benefit of having an executive board chair (relative to another chair type) is lower when board leadership demands are higher. Model 6 in Table 4 finds no support while Models 6 in Tables 5 and 6 provide some support for two-year (p = 0.024) and three-year (p = 0.077) performance, respectively. Again, we plotted the predicted values of two-year ROA for this interaction. As evident in Figure 4, when a given firm's board leadership demands are very great, the performance benefit of an executive board chair position disappears.
Interaction of Executive Board Chair and Board Leadership Demands
Robustness Checks
To further test the robustness of our results, we carried out additional analyses. We explored whether the benefits of the executive board chair position are affected by the presence of a lead independent director (Krause et al., 2017). It might be argued that, as an executive involved in strategy development and implementation, the executive board chair may need to be monitored. Indeed, prior research has found that board chairs who remain involved in firm operations can, at times, hamper performance gains in efforts to maintain control over the firm (Quigley & Hambrick, 2012). In our sample, a lead independent director was present in 53 percent of cases in which an executive board chair was present, compared to 56 percent of cases in which a CEO/chair was present, and 23 percent of cases in which a non-executive chair was present. However, our analysis reveals no interaction effects between the executive board chair and the lead independent director on firm performance. We hesitate to interpret null effects too strongly, but they could potentially imply that executive chairs do not present much of an agency problem, or perhaps that lead independent directors lack sufficient authority to monitor executive chairs.
Given that the performance benefit associated with the executive board chair position is more pronounced when CEO power is higher, we examined whether this effect holds when the firm's organizational complexity or board leadership demands are also higher. We ran three-way interactions of executive board chair, CEO power, and organizational complexity or board leadership demands. We found no effects of the three-way interaction with organizational complexity. However, we found a positive effect of the three-way interaction between executive board chair, CEO power, and board leadership demands on 2-year (p = 0.042) and 3-year (p = 0.034) measures of ROA. Investigation from a split sample analysis offered further support that a given firm's higher performance benefit of executive board chair presence when CEO power is higher is even stronger when board leadership demands are higher. This finding suggests that CEO power might outweigh board leadership demands as a contingency factor for the executive chair relationship with firm performance. While board leadership demands may divide an executive chair's attention as we theorized, the presence of a very powerful CEO may nevertheless make an executive chair worthwhile. Three-way interactions are notoriously difficult to interpret, so we hesitate to make too strong of a claim. However, we do encourage scholars to explore the possibility that counterbalancing CEO power outweighs board leadership as a concern for board chairs, particularly executive chairs, and determine why this might be.
To ensure that the results were not being driven by the effects produced in comparison to only one type of board leadership structure, we ran a series of further analyses including a variable for either CEO duality or a non-executive board chair, effectively testing the executive board chair against each type of alternative board leadership structure. Results were materially unchanged.
To rule out the possibility that our results were being overly driven by instances in which an executive board chair was a founder or another highly influential or powerful owner, we reran the analyses excluding cases in which any board chair had significant ownership of the firm. First, we excluded cases in which the board chair owned more than 10 percent of the firm's shares. Results were materially unchanged. We then lowered the threshold to 5 percent ownership of firm shares. Results also remained materially unchanged.
Our exploratory analyses revealed that executive board chairs are often the firm's former CEO. A new CEO might thus have lower power when an executive board chair is present in a firm. The potentially skewed distribution of CEO power could affect the results from our tests on the interaction of executive board chair presence and CEO power. We found that all levels of CEO power were well represented when an executive chair was present and that the distribution of CEO power when an executive chair was not present was highly similar. Nevertheless, as an extra robustness check, we orthogonalized CEO power from executive chair status and reran our main analyses using these values. Results for this interaction effect remained materially unchanged.
Prior work investigating the influence of a firm's leadership on its performance has suggested that the ROA of financial institutions (SIC codes in the 6000s) or government or unclassified industries (SIC greater than 9000) is not comparable to other industries (Mackey, 2008; McGahan & Porter, 1997; Withers & Fitza, 2017). To ensure that such firms were not confounding our results, we dropped these cases and reran our analyses. Results remained materially unchanged.
In place of ROA, we also ran analyses using Return on Sales (ROS) and Tobin’s Q as alternative measures of performance, again using 1-year, 2-year, and 3-year measures. Results from the full models of these tests are reported in Table 7. As evident in Models 1 to 3, results for Hypotheses 1, 2, and 3 using ROS as a dependent variable were highly consistent with the main results. Results for Hypothesis 4 revealed coefficients in the hypothesized direction but with p values outside standard significance thresholds. As evident in Models 4 to 6, the executive board chair position does not affect a given firm's Tobin's Q overall. However, results from the hypothesized interactions were generally supported, albeit with less robustness than those with the accounting-based performance measures.
Robustness Checks: Second-Stage Fixed Effects OLS Estimates of Other Performance Measures
Note: Robust p values are in parentheses.
Supplementary Analyses
To further probe the performance effect that the executive board chair position offers a given firm, we ran additional analyses. First, unlike the main within-firm analyses that focus on the presence of an executive board chair, we ran between-firm analyses testing the performance effects of the appointment of an executive board chair. To do this, we generated a matched sample of observations by employing propensity score matching to determine the closest match of an observation in which an executive board chair had been appointed (treated) with an observation in which another type of board chair (CEO/chair or non-executive board chair) had been appointed (untreated). These observations were matched from the initial sample of 10,420 observations by several variables calculated in year t-1 that included whether the firm's CEO also held the chair position, the board's independence, the presence of a lead independent director, CEO power, organizational complexity, board leadership demands, TMT power, ROA, and the year of appointment. It also included a variable representing the incoming board chair's ownership of firm stock in year t. The matching strategy required common support and allowed for replacement (i.e., allowed a treated observation to be matched with an untreated observation even if another treated observation had matched with the same untreated observation). We then ran an OLS regression with robust standard errors on this matched sample including the same variables as controls, as well as industry instability and firm R&D intensity, examining whether the appointment of an executive board chair offers a performance benefit to firms. As reported in Models 1 to 3 in Table 8, the appointment of an executive board chair is positively associated with 1-year (p = 0.020), 2-year (p = 0.004), and 3-year (p = 0.000) measures of ROA.
Matched Sample OLS Estimates of Postsuccession Firm Performance: ROA
Note: Robust p values are in parentheses.
Next, we carried out the same tests on a matched sample of observations of non-CEO board chair appointments, effectively testing the performance effects of executive board chair appointments against a matched sample of non-executive board chair appointments. As Models 4 to 6 in Table 8 reveal, we found no support for 1-year ROA, but we did find support for 2-year (p = 0.017) and 3-year (p < 0.000) measures of ROA, respectively. We also carried out the same tests on a matched sample of observations in which the CEO retired to become either an executive board chair or a non-executive board chair. As Models 7 to 9 in Table 8 report, an OLS regression on this matched sample found again no support for 1-year ROA but did find support for 2-year (p = 0.043) and 3-year (p = 0.000) ROA, respectively.
We also ran a between-firm analysis on a subsample of observations that included an executive board chair, comparing the performance benefits of three different types of executive board chairs: founders, family members, and CEOs who retired to the position. Using executive board chairs who were none of the above as the base case, Table 9 reports that executive board chairs who are founders are more positively related to performance. However, retiring CEOs who become executive board chairs do not appear to offer a greater performance benefit, while family members who serve as executive board chairs only show a greater performance advantage for 3-year ROA. Further, a test of coefficients from the three types of executive board chairs included in the models reveals that the performance effect related to a founder is greater than that of a retiring CEO for 1-year (p = 0.006), 2-year (p = 0.012), and 3-year measures of ROA (p = 0.003). The performance effect relative to a family member is not different from a founder and only marginally different from a retiring CEO for 3-year ROA (p = 0.088).
Subsample Analyses of Executive Board Chairs: OLS Estimates of Performance
Note: Robust p values are in parentheses.
Discussion
The executive board chair is a corporate governance hybrid, involved in both strategic decision-making and the oversight of the decisions made. Though the executive board chair position violates the clear separation of decision management from decision control advocated by traditional agency theorists (Fama & Jensen, 1983), it maintains more of a separation than does the common practice of CEO duality while simultaneously contributing insider expertise to the oversight and strategic decision-making processes. As such, the executive board chair position poses a quandary for extant theoretical explanations of board leadership. In this research, we propose that by contributing insider firm knowledge to the board's oversight process, as well as aiding in strategic decision-making, the executive board chair position can benefit firm performance. We also argue that the executive board chair position's performance advantage for a given firm differs depending on how the hybrid nature affects the chair and CEO roles. Our within-firm study of S&P 1500 firms that had an executive board chair at some point between 2003 and 2017 shows that for a given firm the executive board chair position is associated with higher performance than other board chair types, with the advantage being greater when CEO power is higher, and the advantage being smaller when organizational complexity or board leadership demands are higher. Thus, we find strong empirical support for our overall model.
Theoretical Contributions
This research contributes to theory on corporate governance and strategic leadership. We bring some resolution—as well as additional questions—to the debate over the relative merits of insider knowledge and outsider objectivity in board governance (Baysinger & Hoskisson, 1990; Ocasio, 1994; Shen & Cannella, 2002; Zorn et al., 2017). The executive board chair position challenges the predominant view of the governance paradox: that a board with sufficient knowledge of the firm to engage in effective oversight lacks sufficient objectivity, and a sufficiently objective board lacks the necessary knowledge. A hybrid board leadership structure such as the executive chair combines some degree of strategic involvement with some separation from management, thus blurring the lines of the demarcation of roles between the board chair and CEO. By revealing a strong and consistent positive relationship between the executive board chair position and a firm's performance, we provide evidence that the executive board chair may help resolve the trade-off between involvement and independence in governance.
We contribute to the perspective that the board chair acts as a strategic advisor and a board leader. Research on board leadership has focused largely on the mitigation of principal-agent conflict and only recently is it beginning to consider how a separate board chair's resources can be leveraged to aid in strategic decision-making and implementation. Our study suggests that an executive board chair does offer a strategic benefit. However, we also note that there are boundary conditions associated with this and that when a firm is more strategically complex, the extra involvement of an executive board chair may interfere with a CEO's ability to run the firm. This contributes to past work on the limits to a firm's strategic behavior that a non-independent board chair may bring. Indeed, the findings presented in this paper offer further support to Quigley and Hambrick’s (2012) findings that a non-independent board chair's involvement in strategy may have some negative implications. In short, the implications of having a board chair more heavily involved in strategy appear not to be straightforward, suggesting that more research in this area is warranted.
Recent research has also begun to highlight the board chair's role in leading the board and managing its interactions (Bezemer et al., 2018; Veltrop et al., 2021). Our study further demonstrates the importance of this role and that a board chair must maintain focus on these duties. As such, we identify important contingency factors of the relative superiority of the executive board chair position over a non-executive chair and a combined CEO/chair. Our research demonstrates that the executive board chair focuses on strategy and that this may come at the expense of board leadership responsibilities. As research in this domain grows, it will be important for scholars to consider the executive board chair as a unique type of separate board chair with its own set of implications for board leadership.
Beyond these core contributions, our study also adds to recent theoretical developments expanding the construct of board leadership beyond the dichotomous constraints of CEO duality (Krause et al., 2014). Corporate governance scholars have acknowledged that board chairs who are separate from the CEO can play different roles at the firm, with varying effects (Krause, 2017; Krause et al., 2019; Oliver, Krause, Busenbark, & Kalm, 2018; Withers & Fitza, 2017). Others have introduced alternative board leadership structures that have emerged in recent years, such as the lead independent director (Krause et al., 2017; Semadeni & Krause, 2020). We introduce another board leadership innovation that blurs the distinction between independent and non-independent leadership, complicating existing theory on the merits of board independence (Dalton, Daily, Ellstrand, & Johnson, 1998). By doing so, we highlight the interfaces that exist among firms’ strategic leaders (Simsek, Heavey, & Fox, 2018) and explicate some of the implications that these interfaces have for extant governance theory.
Accordingly, our research indicates that the executive board chair offers a strategic benefit, not simply an oversight benefit, thus differentiating itself from a lead independent director. While our analyses find that the presence of a lead independent director does not play a moderating role on the performance effect exhibited by an executive chair, we do believe there is still much work to be done examining these roles and their interdependence. By acknowledging the use of the executive board chair as a hybrid board leadership structure, our research helps the broader corporate governance literature come into closer alignment with the current reality of practice, embracing more of the complexity and idiosyncrasy of firms’ actual governance choices.
Managerial Implications
This study presents several implications for corporate governance practitioners. First and most directly, the research suggests that firms might consider opting for an executive board chair, as doing so is related to a considerable performance advantage over either a non-executive chair or a combined CEO/chair. Moreover, this study also delineates circumstances during which opting for an executive board chair might be particularly advantageous. While firms are continuing to separate the CEO and chair positions in an effort to ensure independent oversight and to avoid the accumulation of too much power in the hands of the CEO, this study suggests that separate—and particularly powerful—CEOs still need to be monitored and that executive board chairs are in a unique position to provide such oversight.
However, firms should also be aware that the presence of an executive board chair blurs the lines of the CEO and chair roles. The increased involvement of an executive board chair may hurt a CEO's unity of direction. While this may mitigate a powerful CEO's ability to unilaterally lead the firm in ways that serve their own interests at the expense of the firm's, it may also complicate strategic decision-making and execution, which as the organization becomes increasingly complex, reduces the performance benefit of having an executive chair. Similarly, an executive board chair's emphasis on strategy may also result in the failure to adequately fulfill the board leadership duties expected of a board chair when such duties are more demanding. While these represent some boundary conditions to the overall benefit of an executive board chair, Figures 3 and 4 demonstrate that having an executive board chair remains beneficial to firm performance for virtually all levels of organizational complexity and board leadership demands. Hence, overall, the advantages of having an executive board chair relative to another board chair type appear to outweigh the disadvantages.
Our analyses also suggest that the executive board chair position may be especially useful for founder-led firms. Eventually, founders must pass the baton. However, the antecedents and consequences of doing so are complex (e.g., Lee, Yoon, & Boivie, 2020). Our analyses support a view held by some, such as Jeff Bezos, the founder of Amazon, who suggested that transitioning to the executive board chair position offers a way to step back from the busy day-to-day operations of the CEO role but still remain closely involved in the firm's efforts to introduce new products and initiatives (Bezos, 2021). Our study offers a first step in considering the role of the executive board chair position in these circumstances and provides founder-led firms something to consider.
The arguments and findings in this study are at odds with some of the business press. While some have suggested that opting for an executive board chair is a bad idea (Hodgson, 2014), this study argues and finds empirical evidence that an executive board chair may indeed be beneficial for some firms. As such, we believe that our study serves as a useful example of where scholarly research can update and inform practitioners and the business press and therefore contributes to the debate among corporate governance pundits about optimal board leadership structures.
Limitations and Extensions
Our study has some limitations that create opportunities for future research. First, we conceptualize the executive board chair and non-executive board chair as distinct and rely on how firms report their board chairs in our measurement of the two. It is possible that some non-executive board chairs are, at times, more involved in strategic decisions—perhaps, for example, during restructuring, mergers, or other punctuated events. It is also possible that some executive board chairs are less involved than some non-executive board chairs. Determining the extent to which either type of board chair is involved in strategic decision making and implementation is virtually impossible. As such, we chose to rely on how firms themselves report the type of board chairs they have. While not perfect, we believe that this is a safer and more conservative measurement than any alternative. That said, we encourage future studies, such as qualitative ones, that might further elucidate how the executive board chair role is conceptualized within the firm.
An assumption of our study is that the presence of an executive board chair is associated with firm performance because of the improved governance and strategic decision-making from better firm-specific knowledge (while being separate from the CEO). Our reasoning follows the well-established agency-theoretic notion that the primary role of the board of directors is to effectively control and monitor the firm's top management. However, we also suggest that the executive board chair is to a considerable extent involved in the actual strategic leadership of the firm, instead of indirectly contributing to firm performance by enhancing control and monitoring, and that their knowledge and experience may also directly benefit firm performance, for example, through strategic insights. It is also important to recognize that different performance measures capture different elements of the executive board chair, and the effects may thus differ depending on the measure. Future research could therefore further disentangle the different board chair roles and their effects by, for example, considering additional outcomes that are possibly closer to the different roles.
In addition, as a hybrid governance phenomenon, the executive board chair may, at times, be a transition role that helps firms moving from one corporate governance model to another. For example, when Alphabet announced the retirement of Eric Schmidt as executive board chair on December 21, 2017, the firm stated that it “anticipates that the Board will appoint a non-executive chairman” (Alphabet, 2017), which they eventually did when they appointed John Hennessy to this position in February 2018. Even though we accounted for potential changes in the firm's governance by controlling for the two other board chair models in the previous year, as well as requiring an individual to remain in the position for at least 1 year to be considered an executive board chair, these longer term transition processes may impact the effectiveness of the executive chair position. Similarly, this paper does not consider how the benefits of an executive board chair may change over time or whether the executive board chair may grow “stale in the saddle” or obsolete. Indeed, the benefits of a strategic leader may decrease as the leader's fit with the contextual needs of the firm changes (e.g., Chen & Hambrick, 2012). While this is an important question to consider, our focus was on examining whether a given firm will have, on average, a higher performance with the executive board chair than with other board chair types. Examining whether an executive board chair offers the greatest benefit following a succession, or for a certain period of time, is an important question that deserves investigation. We therefore see an opportunity for future research that explores the transition processes between different board leadership structures and their effectiveness over time (see Gove & Junkunc, 2013).
Conclusion
The executive board chair is a corporate governance hybrid, involved in both strategic decision-making and the oversight of strategy implementation. In this study, we argue that the executive board chair may offer a firm performance advantage by contributing the necessary expertise for effective firm governance and strategic decision-making. In addition, we posit that the executive board chair may allow a firm to more effectively engage in oversight, especially when the CEO is powerful, but that they will be less effective when organizational complexity or board leadership demands are higher. Since our analyses support our arguments, we hope that this study will help guide future research on the role of the board chair.
Footnotes
Acknowledgments
We thank Senior Associate Editor Zeki Simsek and the anonymous reviewers for their efforts in the development of the paper. We also thank John Busenbark and Matthew Semadeni for their helpful advice. We gratefully acknowledge project funding from the Swiss National Science Foundation (grant number 185173).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Swiss National Science Foundation (grant number 185173).
